Emotions from Text: Machine Learning for Text-based Emotion Prediction

By Cecilia Ovesdotter Alm, Dan Roth, Richard Sproat. Human Language Technology - The Baltic Perspectiv, 2005

999 citations recorded by the source corpus.

In addition to information, text contains attitudinal, and more specifically, emotional content. This paper explores the text-based emotion prediction problem empirically, using supervised machine learning with the SNoW learning architecture. The goal is to classify the emotional affinity of sentences in the narrative domain of children's fairy tales, for subsequent usage in appropriate expressive rendering of text-to-speech synthesis. Initial experiments on a preliminary data set of 22 fairy tales show encouraging results over a naive baseline and BOW approach for classification of emotional versus non-emotional contents, with some dependency on parameter tuning. We also discuss results for a tripartite model which covers emotional valence, as well as feature set alternations. In addition, we present plans for a more cognitively sound sequential model, taking into consideration a larger set of basic emotions.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/872352b0a53ab6cbb4420f81df64d215d86c7d9b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Sentiment Analysis (2025, neutral): “• Extended to different sentiment dimensions and different categories sets (Cabral and Hortaçsu 2006; Pang and Lee 2005; Goldberg and Zhu 2006; Snyder and Barzilay 2007; Bruce and Wiebe 1999; Wiebe et al. 1999; Hatzivassiloglou and Wiebe 2000; Riloff et al. 2005; Wiebe et al. 2005; Pang and Lee 2004”
Identifying Expressions of Emotion in Text (2007, background): “[1] have explored automatic classification of sentences in children's fairy tales according to the basic emotions identified by Ekman [3].”
Opinion Mining and Sentiment Analysis (2008, background): “Researchers have considered various affect types, such as the six “universal” emotions [86]: anger, disgust, fear, happiness, sadness, and surprise [9, 192, 286].”
Sentiment Analysis: Detecting Valence, Emotions, and Other Affectual States from Text (2016, background): “Alm, Roth, and Sproat (2005) annotated 22 Grimm fairy tales (1580 sentences) for Ekman emotions.15 Strapparava and Mihalcea (2007) annotated newspaper headlines with intensity scores for each of the Ekman emotions, referred to as the Text Affect Dataset.16 Aman and Szpakowicz (2007) annotated blog…”
The Hourglass of Emotions (2011, background): “Several affect recognition and sentiment analysis systems [38–44] are based on different emotion categorisation models, which generally comprise a relatively small set of categories (Table 2).”
#Emotional Tweets (2012, background): “Alm et al. (2005) and Francisco and Gervás (2006) worked on fairy tales. Boucouvalas (2002), John et al. (2006), and Mohammad (2012a) explored emotions in novels. Zhe and Boucouvalas (2002), Holzman and Pottenger (2003), and Ma et al. (2005) annotated chat messages for emotions. Liu et al. (2003) an”
ICON: Interactive Conversational Memory Network for Multimodal Emotion Detection (2018, background): “The role of text in emotional analysis became evident with later research such as Alm et al. (2005); Strapparava and Mi-halcea (2010).”
Emotional state classification from EEG data using machine learning approach (2014, background): “Many different materials have been used to elicit emotions in the laboratory such as facial expressions [22], slides [23], texts [24], music [25], and movies [26].”
Learning Sentiment Lexicons in Spanish (2012, methodology): “…applications have already used techniques for automatic sentiment and sub-jectivity analysis, including expressive text-to-speech synthesis (Alm et al., 2005), tracking sentiment timelines in on-line forums and news (Lloyd et al., 2005; Balog et al., 2006), analysis of political debates…”
Fusing audio, visual and textual clues for sentiment analysis from multimodal content (2016, background): “The task of automatically identifying fine grained emotions, such as anger, joy, surprise, fear, disgust, and sadness, explicitly or implicitly expressed in a text has been addressed by several researchers [36,37].”
Learning Word Vectors for Sentiment Analysis (2011, methodology): “Thus, we extend the model with a supervised sentiment component that is capable of embracing many social and attitudinal aspects of meaning (Wilson et al., 2004; Alm et al., 2005; Andreevskaia and Bergler, 2006; Pang and Lee, 2005; Goldberg and Zhu, 2006; Snyder and Barzilay, 2007).”
GoEmotions: A Dataset of Fine-Grained Emotions (2020, background): “Other datasets annotate news headlines (Strapparava and Mihalcea, 2007), dialogs (Li et al., 2017), fairy-tales (Alm et al., 2005), movie subtitles ( ¨Ohman et al., 2018), sentences based on FrameNet (Ghazi et al., 2015), or self-reported experiences (Scherer and Wallbott, 1994) among other domains.”
SemEval-2019 Task 3: EmoContext Contextual Emotion Detection in Text (2019, methodology): “…of sentences which are 3 Task webpage: humanizing-ai.com/emocontext.html then used as input by classiﬁers such as Decision Trees, SVMs among others to predict the output (Alm et al., 2005), (Balabantaray et al., 2012), (Davidov et al., 2010), (Kunneman et al., 2014), (Yan and Turtle, 2016).”